Orca-4B

Orca: The World is in Your Mind Orca Team, Beijing Academy of Artificial Intelligence ⭐️ Project Page  |  🤗 Hugging Face  |  📑 Technical Report A general world foundation model centered on Next-State-Prediction. 💬 If you have any questions, feel free to contact us via WeChat. 🔥 Overview Orca** is an initial instantiation of a general world foundation model. It learns a unified world latent space from multimodal world signals and exposes the learned latent through multimodal readout interfaces. Rather than optimizing isolated **next-token**, **next-frame**, or **next-action** prediction objectives, Orca is centered on **Next-State-Prediction**: a unified state-transition modeling route toward understanding, predicting, and acting upon the world. In this version, Orca focuses on two fundamental input signals: **visual signals** for dense observations of world evolution, and **language signals** for event descriptions, task intentions, causal explanations, and semantic constraints. **Unconscious learning**: dense natural transitions from continuous videos. **Conscious learning**: sparse meaningful transitions under language-described events and VQA supervision. **Frozen-backbone readouts**: lightweight decoders for **text**, **images**, and **actions**. **Scaling analysis**: stronger world modeling, stronger downstream readouts. 🗞️ News **2026-07-14**: 🚀 **Orca-4B** checkpoint was released on HuggingFace. **`2026-06-29`**: 🎉 **Orca Technical Report** was released. 📆 Todo [x] Release the **Orca Technical Report**. [x] Release the **Orca-4B checkpoint** for world latent learning and downstream readouts. [x] Release **inference code** for text, image, and action readouts. [ ] Release the **Orca-0.8B checkpoint** for lightweight research and reproduction. [ ] Release **downstream fine-tuning code** for modality-specific readout adaptation. ⭐️ Architecture Orca follows an **Encoder-Decoder** architecture. Given multimodal world signals, the **Encoder** l